Talk:Integral/Archive 4

The history section needs a careful vetting. It retains the basic outlines given to it by the creator of the section, and appears to give a fairly subjective description of the history of integration, to the point of OR. For example, while there is no doubt that ancient Egyptians asked, and sometimes answered, questions about areas and volumes, is it commonly considered to be "integration"? Likewise, it's better not to go into priority questions (Gregory vs Newton vs Leibniz), and refer to other articles for the fine details of FTC, invention of calculus, and so on. Since this is an article about integral, and not these other subjects, we can afford doing so! Arcfrk 23:22, 8 August 2007 (UTC)

I wrote the first deaft of the history section. The Egyptian stuff can potentially stay; it's only a sentence that was cribbed from History of Calculus. Certainly the material on the method of exhaustion should be there. Wrt the stuff about Gregory -- I'm not sure when that slipped in there, but its rather undue weight to the topic. I believe Barrow is also cited as having had some early insight, but again, its undue weight to get into such hair splitting here -- it can go in History of Calculus if it needs to go anywhere. -- Leland McInnes 13:25, 9 August 2007 (UTC)

Just to clarify: by "vetting" I mean checking the text against printed authoritative sources on history of mathematics, not against other wikipedia articles or MacTutor and other compilatory web resources. Besides obvious problems with circularity, the present quality of scholarship even at the better websites is only in a mediocre to fair range. Arcfrk 18:28, 9 August 2007 (UTC)

The sentence about Gregory is already in History of Calculus, right in the lead of that article. This is why I copied it in this article. If Gregory and Barrow contributed, I suggest to add a short sentence about that. It is precious epistemologic information. When written properly, history tells people that masters are not gods, and that science does not come from heaven, but always builds from previous knowledge. Popularizing this concept, even is short summaries, is an important mission, rather than hair splitting. And it doesn't require a lot of space. A short sentence is nothing compared to the length of the history section. Paolo.dL 08:57, 10 August 2007 (UTC)

You raise interesting points, but here are a couple of remarks.

• The article History of calculus does not have a lead. The historical section, indeed, mentions Gregory (although I needed to use the "find" function of my browser, it's not by any means a highly visible reference), but then says

Of course, important contributions were also made by Barrow, Descartes, de Fermat, Huygens, Wallis and many others.

Consequently, singling out Gregory is giving "undue weight" that Leland McInnes mentioned above. This is a somewhat subtler issue than just the length of the lead.

• You had not read what appears right above your comment before posting it, or else the meaning completely escaped you. Just because a statement appears in another wikipedia article (quarternary source), it needs not be true, accurate, universally accepted, etc. As the article develops, it's inevitable that there would be a lot of unsourced material put in it; however, at some point it becomes necessary to verify these statements with primary and secondary sources. Arcfrk 18:20, 10 August 2007 (UTC)

Being picky :-) How comes it was so difficult for you to find the sentence about Gregory? :-) Yes, the article about history does not have a lead but it has a leading section, and in this leading section, the leading paragraph (not only sentence, but a separate paragraph) of the subsection about "Modern calculus" is fully dedicated to Gregory!

Secondary source. More importantly, the bibliographic reference about Gregory, which I have included in my edit and you possibly have not seen, was given in the lead of the article about the Fundamental Theorem of Calculus, where again a whole (short) paragraph was dedicated to Gregory (by the way, in that article Newton and Leibniz are not even cited, and I will insert there a citation ASAP). So, I did not ignore your comment about the quality of other articles in Wikipedia. Of course, the authority of the authors of the book (Marlow Anderson, Robin J. Wilson, Victor J. Katz) can be denied, but they happen to be university professors. Also, they provide a rich list of primary references. I think that you should also trust the people who wrote the history of calculus section in Wikipedia. They are as good as you are. But I am sure you agree that if you don't trust them, then you should provide another reference against their claim, show that your reference is more reliable, and delete their comment about Gregor in the other two articles. Coherence in these three articles is important.

Undue weight: true! Now, the important question is: didn't my statement about epistemology and popularization touch you? This should lead you to consider that there is even now an "undue weight" given to Newton and Leibniz. So, wouldn't it be nicer to just start from what I did, refine it if needed, change the subheader, and add at the end of the subsection the short sentence about Barrow, Descartes, de Fermat, Huygens, Wallis and many others, rather than totally rejecting my contribution?

Less unbalanced subheader. For instance, the subheader "Modern calculus" would be less unbalanced than "Newton and Leibniz" (curent version) and "Gregory, Newton and Leibniz" (my previous suggestion).

In the meantime, thanks a lot Arcfrk for your interesting comment. And, for all of you: remember that I appreciate your work and I will always respect your final decision. With kind regards, Paolo.dL 21:16, 10 August 2007 (UTC)

I've got a very busy day, but I will take a few minutes to outline some of the many problems with the history section at large, and Paolo's edit in specific.

The latter first, concentrating on this difference. To make discussion easier, here is Paolo's version:

---

Gregory, Newton, and Leibniz

The major advance in integration came in the 17th Century. The first published statement and proof of a restricted version of the fundamental theorem of calculus was by James Gregory (1638-1675)^[1]. Isaac Newton (1643–1727) and Gottfried Leibniz (1646–1716) independently developed the theorem in its final form. The theorem demonstrates a connection between integration and differentiation. This connection, combined with the comparative ease of differentiation, can be exploited to calculate integrals. In particular, the fundamental theorem of calculus allows one to solve a much broader class of problems. Equal in importance is the comprehensive mathematical framework that both Newton and Leibniz developed. Given the name infinitesimal calculus, it allowed for precise analysis of functions within continuous domains. This framework eventually became modern Calculus, whose notation for integrals is drawn directly from the work of Leibniz.

…

Notes

1. [1]

   See, e.g., Marlow Anderson, Victor J. Katz, Robin J. Wilson, Sherlock Holmes in Babylon and Other Tales of Mathematical History, Mathematical Association of America, 2004, p. 114.

---

Here are some of the mistakes:

1. It is far more important to state that the advance was the fundamental theorem than to know who first stated it.
2. The date range for Gregory misuses a hyphen instead of the proper en dash.
3. We do not use chatty footnotes in this article; we use Harvard-style references.
4. The question of priorities in calculus is a well-known source of contention, and this edit only adds fuel to the fire.

In fact, the title of the section might more properly be "The fundamental theorem"; in that regard Paolo's edit exacerbated a prior problem, giving far too much importance to Gregory, as others have pointed out.

Let us now turn to the larger problem. I am willing to accept the history section as a first draft, but I agree with Arcfrk that more care is required, and that most of this — including the thorny issues of priority — is best left to the separate article. As it stands, it does not meet proper standards of scholarship and referencing for its account. Already in the first section there is no mention that Archimedes calculated the area and volume of a sphere, his proudest accomplishment; and he determined the area inside a circle exactly as πR², using that to approximate π; and recent work on the Archimedes Palimpsest has revealed a partial copy of his long-lost "Method", making it clear he knew far more than previously realized. That is just one example!

Still, the tripartite organization is nearly sound. We might label these the geometric phase, the algebraic phase, and the analytic phase. Omitted is a more recent phase I have no catchy name for, with ideas like smooth infinitesimal analysis. --KSmrq^T 20:53, 10 August 2007 (UTC)

Interesting. Thanks a lot for explaining so clearly. I sincerely appreciate your patience. I already wrote my opinion in detail above, in my answer to Arcfrk, which I edited before reading your comment. A few notes:

• I agree about point 1, and of course 2 and 3. The next notes are about point 4.
• I am convinced that Newton and Leibniz are the great masters of the 17th century, but did not come from heaven; the important news which is missing in the curent subsection is that there was also a group of secondary authors in that period who contributed to the developement of modern calculus; this is what I enjoied to learn.
• As a reader, I greatly appreciated the boldness and the invaluable contribution of the editor who gave some (limited but important) credit to these secondary authors on Wikipedia.
• Where's our boldness? The fire can be easily controlled:

1. by proper hedging;
2. by the fact that we provide a good secondary reference, that everybody can read on line. Those who possibly (but unlikely) will want to counterclaim will need to give another reference.

• Indeed, nobody objected and no fire was lit in the other two talk pages, mainly because it was clearly stated that Gregory did not create the theorem: he just studied a special case of its second part (similarly, although Cardano cannot be compared to Newton and Leibniz, in his famous book Ars Magna, where he provided his general solution of the third degree equations, he gave credit to Tartaglia for the solution of a special case).

Thanks again. With kind regards, Paolo.dL 22:04, 10 August 2007 (UTC)

Forgive me if I'm wrong, but wasn't the fundamental theorem of calculus (which by the way, was better-written and more accurate than the one on that page itself) specifically requested for this article? I spent a while getting a completely correct statement down as generally requested. I understand that the article is long, but having an article on integration without this was deemed incomplete. And now it's no longer there. Thoughts? Xantharius 15:43, 9 August 2007 (UTC)

Yes, I'm also opposed to just delete the section on the f.t.c. Hey, what does the layman know about integrals (if anything): this theorem. Just mentioning it with one word in the lead is imo not enough. Almost every topic in this deserves (and has) its separate article, which is no reason to simply delete these sections. Jakob.scholbach 15:54, 9 August 2007 (UTC)

It would appear that we should put it back in. I will unless there are objections (but remember, to have the FTOC on this page was the general inclination for a long time . . .) Xantharius 15:57, 9 August 2007 (UTC)

Yes to the above remarks. The section on the Fundamental Theorem of Calculus is (forgive me for a pun) integral to this article; it was not overly long, having been written in the summary format, and needs to be restored. Arcfrk 18:21, 9 August 2007 (UTC)

I agree. The section about the Fundamental Theorem was good and tight. It should be in here. I'm also asking Paolo.dL to please discuss changes he is proposing on this talk page. DavidCBryant 19:51, 9 August 2007 (UTC)

Thanks, David. You beat me to it (inserting the section on the FTOC) by about 10 seconds! Xantharius 19:59, 9 August 2007 (UTC)

I respect the opinion of those who so skillfully wrote this complex article, and I greatly appreciate your polite explanations (and special thanks to David for warning me on my talk page; I had forgot to insert this article on my watchlist). The final decision is yours. But let me try and see if I can convince you. The first paragraph of the "Methods and applications" section states "The most basic technique for computing integrals of one real variable is based on the fundamental theorem of calculus. It proceeds like this:", and then explains the second theorem, and includes a copy of the formula. I thought that this explanation was enough. For details, the reader can use the internal link, which is provided right when needed. Paolo.dL 07:31, 10 August 2007 (UTC)

The reason why the Theorem is discussed before the "Extensions" and after the "Properties" is not clear. If you decide to keep the section about the Theorem, wouldn't it be better to move it right above the section on "Methods and applications"? The word theorem is used only in the lead and in the "Methods and applications" section. Paolo.dL 07:31, 10 August 2007 (UTC)

Well, I find it pretty reasonable there. IMO it should be 1. Intro, 2. Notation 3. History, 4. Formal Defs, 5. Properties, 6. FTOC and so on. It seems to me to be the first non-trivial theorem about the integral which appears in the text. The mentioned extensions and methods are IMO something which should (for intrinsic reasons) follow these things (as it is now).

Another remark concerning your dispute with KSmrq: During the long and winding process of improving this article it has become somewhat standard to most people around to 1. think thrice before editing, and 2. if doing major edits, explain this using the talk page.

Ad 1. The article is a subtle web of intertwining facts, ideas etc. So, for example, in one of your edits you wrote

... is called domain of integration or interval of integration. The domain is defined by two values referred to as the lower limit and upper limit of integration.

This line is totally OK if you don't read any further. But, if you look down, the article talks about integration on manifolds, on measure spaces. So, at this point it is misleading to talk about lower and upper limits of the domain. Edits of this kind were done in good will, I'm sure, but the result may be counterproductive. Please don't hesitate to improve the article further (with equal emphasis on both words). Having done some dozens of edits here is no justification of "owning" the article, surely, but it gets tiring if one needs to correct things which one has thought about carefully. This may have been KSmrq's situation with your edits.

Ad 2: see for example my edit above Talk:Integral#polished_introduction, which was controversial, too, but instead of being reverted it got a (not yet finished, it seems) constructive discussion. So, discussing it here may take some more time first, but it will save time later, for everybody. Jakob.scholbach 15:47, 10 August 2007 (UTC)

Thank you very much for explaining. About the lower limit and upper limit, it appeared to me a very conservative edit. I now see why you didn't like it. That teaches me a lot, and confirms how much accurate was the work done by your group. KSmrq just reverted my edits without explaining why. The consequence of his behaviour was that, before reading your contribution, I had not understood my mistake yet and wouldn't have been able to avoid similar mistakes in the future.

By the way 1, my edit about the limits of integraton was coherent with what I have read elsewhere; you are right that it was not perfectly correct in the general case, but then you should agree that also the sentence about history in the lead is not perfectly correct and not coherent with what is written right below, in the ensuing section, as I maintained above... either we are both splitting hair, or, as I see it, we are both trying to do our job with care :-)

Minor edits. I always think trice before editing, and endeavour to respect as much as possible the terminology and the ideas of the authors, even though this does not prevent me from making mistakes. I thought that my edits on this article were conservative enough not to deserve a discussion. I was convinced that a well written edit summary was sufficient. For instance, my edit about Gregory in the history section (see discussion above) was just for coherence with the introduction of the History of calculus article. I just copied a sentence from there, together with the reference supporting it, and adjusted it to fit in the new context. Except for the deletion of the section about the Theorem, for which I apologize, I still believe that the other edits were minor and conservative.

Major edits. On second thought, I now agree that deleting the section about the Theorem was a major edit which I shouldn't have done without discussing it here. However, in other pages (for instance Talk:Cross product, Talk:Antiderivatives, Talk:Fundamental Theorem of Calculus) you can see that I am used to discuss major edits.

By the way 2, I wholeheartedly agree with Xantharius that the section about the Theorem "was better-written and more accurate" than the one on the separate article, when it was written. In fact, I edited that article before editing this one. I extensively edited the structure and contents of its lead, corrected a questionable sentence at the beginning of the "Formal statement" section, and improved the formal statement itself. I am very happy of all the changes I did in that article, which I partly explained on the talk page, and which in my opinion made the article more accurate and at the same time more synthetic and effective. That's one of the reasons why I thought the short summary in this article was not needed anymore.

So, don't worry, please. I hope you have enough information to rest sure that I am a respectful editor. And you should know that just a hint of what you did, kindly provided by David on my talk page, was enough to make me even more respectful! Paolo.dL 19:49, 10 August 2007 (UTC)

Now that the article has stabilized a little, perhaps it's time to think about the Introduction. One of the first things that I feel should happen is the elimination of the plural first person and the conversational tone, as per the Wikipedia Mathematical MOS. This can be a hard thing to do, though. Also, I'm not sure about being asked to consider a swimming pool. What if I don't want to? (Half-joking.) Further, this is a good analogy, perhaps, but swimming pools often just have linear slopes, which therefore don't require calculus to find their volumes.

Perhaps we can get some consensus before going to town on this, and then do and evaluate. Xantharius 18:48, 10 August 2007 (UTC)

Swimming pools usually have bottoms rounded at the edges, and can have interesting shapes. ;-)

As for the state of the article, I would say it has stagnated rather than stabilized. A number of planned improvements have not happened. --KSmrq^T 07:34, 14 August 2007 (UTC)

Would it be simpler if the concrete function sqrt(x) would be replaced by a function "f", just abstractly? This would make obsolete (and impossible) the actual computation of the approximations, which was criticized above by KSmrq and Leland McInnes. I don't know whether their silence above means concurrence on the matters discussed there. At least, a concrete example of an anti-derivative should be in the intro.

Instead of the pool, we can take a lake, they are hardly ever linear. Also possible is a cup, but the top-surface is usually round. Let's not take a paper mug, though :-). The cup has the advantage that one could rediscuss it in the sections on surface and volume integration, because there are formulae for such surfaces of revolution. Jakob.scholbach 21:32, 10 August 2007 (UTC)

By the way, could somebody archive some parts of the talk page? I don't know how to do this. Jakob.scholbach 21:33, 10 August 2007 (UTC)

Done. DavidCBryant 04:06, 12 August 2007 (UTC)

My silence has more been to do with lack of time. I keep meanin to come back and give the intro a serious going over, but I am frantically busy with far too many things right now to get involved in the protracted debates that will ensue. Instead I will continue to drop by occasionally and try and make my opinion heard so that changes can hopefully incorporate some of the things I feel are important for the introduction. Here are a few points:

• We often get complaints about "unreadable" math articles filled with equations and gobbledegook and no simple clear English explanations. The introduction is, and should be, the place where we provide the simple clear English explanation (people who read sections called "formal definition" deserve what they get in the way of equations and gobbledegook). That means we should strive to not have any equations in the introduction. That is, of course, a hard task, but I think it is entirely doable provising we think carefully about how to explain things beforehand.
• I'm not sure I like the idea of starting with geometry as the example, as it tends to imply that integration is about geometry. Rather, integration is about extending discrete summation into the realm of the continuous -- indeed, calculus in general is about straddling that divide between discrete and continuous -- the fact that it is so applicable to geometry is because geometry deals with continuous worlds, and thus a theory of the continuous is required. Integration is half of that theory (in a sense). We should thus be focussing on the discrete/continuous dichotomy rather than arbitrary geometric examples. That integration is a kind of continuous summation is ultimately, I believe, the core idea that we need the introduction to express.
• Worked examples should be right out. They really shouldn't be necessary. In doing that we are explaining the mechnics of integration, rather than explaining what it is.

I'll continue trying to stop by and dropping in comments when I can. -- Leland McInnes 00:37, 12 August 2007 (UTC)

Why is the introduction restricting line integrals to two to three dimensional functions? Rubybrian (talk) 14:19, 10 December 2007 (UTC)

Integral is in my view getting quite close to successful WP:GA nomination. The coverage of topics is (at least) sufficient at the moment, and the lead is certainly among the best-developed and accessible among maths articles.

However, I feel there are still certain issues with the overall structure and the flow of the article, issues that have to do with the organisation of mathematical content. I also feel some of these issues would be easier to resolve if the more specialised articles referenced from / referencing to Integral were in better shape and formed a more coherent whole. I have added a preliminary proposal for work on these relating articles to better enable Integral to stay focussed.

I have collected below some obervations and proposals specific to the article Integral itself.

The article mixes different types of integral in a way which is likely confusing to a beginner. There are in a way two broad topics within the scope of the article:

1. Integration of functions: This is where belong the successive generesations of functions that can be integrated, the end point of which is (essentially) the measure-theoretic integral. This has several themes with sequences of (logical and historical) steps:
   • Integration of more and more "irregular" functions: from Riemann integral to Lebesgue integral (with respect to the Lebesgue measure) on R
   • Integration "with weights": from Riemann-Stieltjes integral to general measures (and Radon-Nikodym)
   • Integration on more general domains: multiple integrals, then eventually general measurable spaces (with Fubini's theorem an important ingredient)
   • Integration of complex-valued, finite-dimensional vector-valued and general vector-valued functions; vectorial measures
2. Integration of differential forms (from the simple cases such as line integrals to the general case).

As it stands, the article introduces (without much discussion) the general measure-theoretic integral (calling it Lebesgue integral) quite early. However, after that, the unifying nature (among the topics of type 1 above) of the general integral is largely lost, and various different integrals (Riemann-Stieltjes, Lebesgue-Stieltjes, Daniell) are introduced/mentioned without making the connections clear. One thing I'm afraid can easily happen is that the (general) Lebesgue integral gets confused with integration with respect to the Lebesgue measure. In similar vein, multiple integral (Fubini's theorem, really) is introduced as a way to move beyond integrands defined on more general domains than intervals in R, despite the fact that the genereal integral has already been defined. The "Properties of integral" section discusses linearity three times in increasing generality, then discusses various inequalities that hold for the general integral in terms of the Riemann integral. Finally, the "Extensions" section introduces three quite different concepts on the same level: improper integrals, multiple integrals and (the various types of) integrals of differential forms. Multiple integrals should probably appear earlier, while the integrals of differential forms should really be its own top-level section (preferably a short one that points to another article).

"2 Terminology and notation" and "5.3 Conventions" should probably be integrated with other sections, and Conventions made much shorter.

The History section should cite a few sources. Good ones (also for some expansion) are the Historical Notes in Bourbaki's Functions of a Real Variable (pp. 129-162 of the English edition) and Integration (pp. V.123-136). History should probably be moved later in the article.

Most of what I would suggest doing to the article has to do with restructuring and moving text around, condencing where possible (hopefully leveraging related articles brought in line with this one). Stca74 13:09, 9 September 2007 (UTC)

Added vector-valued integrals that were missing in my list above. We cannot realistically plan for any deep treatement here, but should at least cover complex-valued integration and provide pointers to more general set-up Stca74 07:03, 10 September 2007 (UTC)

Most of what you say all makes good sense to me. I will try to help if and when I cna find some free time to work on such things. -- Leland McInnes 15:35, 9 September 2007 (UTC)

Question?? 1)why we are using integration??can you give real time examples?? 2)where does it come to the picture?? —Preceding unsigned comment added by 60.254.13.8 (talk) 18:21, 19 September 2007 (UTC)

Skand swarup (talk) 13:44, 8 May 2008 (UTC) Integration is used to find areas of figures which are not geometric. Suppose you spill water on the floor and want to find out what area the water has covered, you can do so by integrtion. What it does is that it breaks up the non-geometric shape into a number of tiny geometric shapes. It then calculates the area of each of the tiny figures and adds them up. This of course gives only an approximation to the actual area.

Leland McInnes reverted my attempt to make the article more accessible with the comment

rv: It was well intentioned, but really not that helpful, nor all that accurate.

With respect I believe you need to offer more of an explanation. The text I appended to the intro paragraph was the following.

Loosely speaking, the integral can be thought of as the total sum of the output of a process over some period. For example, the integral of the productivity of a factory (e.g. how much the factory produces per hour) over the course of a day would be the total amount the factory produced that day.

To be frank the article as it stands is a bit "gear-headed". That is, mostly the only people who would understand it would be those who already have a background in calculus. In principle since an encyclopedia is supposed to be generally accessible (Wikipedia even more than most) it is worth at least trying to make the introduction clarify the subject matter for as wide an audience as possible (I'd argue that the goal should be broader than that but at least that is a starting point).

--Mcorazao 21:52, 12 October 2007 (UTC)

Please comment or I will simply put back my edits. --Mcorazao 15:15, 16 October 2007 (UTC)

Yes, I will comment. I agree with Leland McInnes - I applaud your intentions but I don't think your example is helpful. You wouldn't integrate hourly production to get the daily production - you would just add up the hourly producton values. You would only integrate if you had an instantaneous productvity rate - which begs the question of how you would measure that. And productivity can only be positive, not negative. Perhaps you can come up with a simpler example ? Gandalf61 20:15, 16 October 2007 (UTC)

I was making links for my page direct integration of a beam and when I was putting in the link for this page, I noticed that the lead for the article reinforces the perception that an integral only represents "the area under a curve." Since there's not another page on mathematical integration, this is more of an issue in this article. I would think the lead should point out to readers that what f(x)dx represents is a box with width dx and height f(x), and that the integral of it is a summation of these rectangles' areas as dx->0 while x goes from a to b. JW 05:44, 13 October 2007 (UTC)

Your idea about a box is not entirely new and not entirely correct. If f(x)<0 then the height would be negative, which it can not be. dx is the differential and can not be width of anything. (Igny 19:09, 13 October 2007 (UTC))

seeing the wide variety of meanings at integral (disambiguation), and the status of the word integral as a generic English word, I wonder whether it wouldn't be advisable to move this article to integral (calculus) on grounds of the "principle of least surprise". dab (�) 18:21, 13 November 2007 (UTC)

Almost all those meanings involve the word "integral" in the role of an adjective, hence qualifying another subject. Used as an independent subject, "integral" fairly unambiguously refers to the notion of integral from calculus and (in a more advanced form) analysis. Arcfrk 20:05, 13 November 2007 (UTC)

it is true that the word is both an adjective and a noun. The OED for the noun has the following:

1. Something entire or undivided; a whole, either as wanting no part, or as made up of parts (Obs. exc. as transf. from 4 (math.) = total sum.)

2. An integral part or element; a constituent, component (Obs.)

3. Gram. Applied by Wilkins to those words or parts of speech which of themselves express a distinct notion, as distinct from those which express relations between notions. Obs.

4. Math. (this article)

so, yes, the noun is obsolete except for its mathematical meaning. The adjective otoh is perfectly current. dab (�) 08:07, 14 November 2007 (UTC)

Do we frequently have articles/dab for the sake of on non-noun alternate usages? Adjectives seem to be our sister project's territory. In any case, looking at the dab page, I don't think anyone will be surprised to end up here after typing integral into the search engine. —Cronholm¹⁴⁴ 08:18, 14 November 2007 (UTC)

you may be right. I was thinking of the possibility that people may want to link to the adjective integral in articles on philosophical topics (e.g., in the context of Integralism) without expecting to link to a calculus topic, but that's something that will just have to be fixed as it occurs. thanks, dab (�) 12:47, 16 November 2007 (UTC)

A tag has been placed on the article. I am tempted to just remove it whilst citing Wikipedia:Scientific citation guidelines, but I think that a discussion should take place here first for the sake of completeness. Thoughts? —Cronholm¹⁴⁴ 05:39, 13 December 2007 (UTC)

It seems to me that the animation associated with the line integral section is incorrect or at least misleading (I feel incorrect). It does give a good graphical explanation of a mathematical operation; however, the operation shown is not a line integral. As I understand it, a line integral on a vector field returns a scalar value; the 'sum' of the dot products between the unit tangent vector to the curve and the field value at each point. However, the graphic shown just gives the 'sum' of the field values (not dot products) (and is vector valued).

I am only a second year degree student, so it is quite possible there are other definitions of line integration I don't know of, or that I am just incorrect. However, being a degree student I have been directly affected by the animation; not knowing the definition of a line integral, I used the animation as a definition in a set problem (which caused problems for me!). I can say from experience therefore that the animation was misleading and definately unhelpful. If it is a (rather than the) correct definition, a note to this effect would be very useful.

If I don't find a reply or changes to the article, I will remove the image to avoid other people getting the same problems from it I did. I'm also posting this message in the main Line Integral article, which uses the same image. —Preceding unsigned comment added by 88.106.245.46 (talk) 12:06, 25 December 2007 (UTC)

Line integrals can have vector values. That the particular line integral the previous editor was calculating did not immediately lend itself to assistance from the supplied animation does not make the animation misleading. As such, I don't think that there is that serious a problem with the image to warrant removing it completely, without providing a replacement animation which addresses the issues that the previous editor raised. (Besides, I thought the animation was really good.) Xantharius (talk) 10:43, 28 December 2007 (UTC)

I don't think that this particular animation adds anything of value. On the other hand, yes, it is very confusing, and should better be removed. Having said that, I am truly amazed that anyone would try to infer a basic mathematical definition from an illustration in the middle of a wikipedia article! Arcfrk (talk) 17:59, 29 December 2007 (UTC)

This article reads too much like a text book for trained mathematicians. It is largely inaccessible to lay-people in search of general and simplified knowledge of the subject. I found it disappointingly unhelpful. 143.97.2.35 (talk) 16:07, 27 December 2007 (UTC)

What were you hoping for? Why do you want to know about integration? The article does need some mathematical knowledge to understand, but there is a limit to how simple it can be reasonably made. Do you need to know what integration in general is, how to do it, or about a specific type of integration (line integration for example)? If you are more specific I will try to improve the article.

The juxtaposition of two symbols have different meanings.

f(x) is the value of function f at argument x.

dx is the differential of a variable x.

f(x)dx is the product of these two.

This is hard to newcomers. I suggest that the multiplication sign be written explicitely in order to reduce the confusion.

Write

${\displaystyle \int f(x)\cdot dx}$

rather than

${\displaystyle \int f(x)dx}$

Any objections? Bo Jacoby (talk) 22:55, 9 January 2008 (UTC)

Dear Bo,

I'm not sure that I like the idea of ${\displaystyle f(x)\cdot dx}$ because I think it might mislead a reader into thinking that ${\displaystyle dx}$ is a number. It is useful to emphasize that ${\displaystyle dx}$ is not a product of ${\displaystyle d}$ and ${\displaystyle x}$, which is why one usually puts a \, between the ${\displaystyle f(x)}$ and ${\displaystyle dx}$, like so:

${\displaystyle \int f(x)\,dx}$.

Sincerely,

Loisel (talk) 03:34, 10 January 2008 (UTC)

I think it is a poor idea, since interpreting f(x)dx as the product of f(x) and dx isn't necessarily correct and might lead one to believe dx actually represented an infintesimal, which is certainly quite wrong for many (or indeed almost all) defintions of integral. Ultimately for many definitions the dx is purely formal notation and not really representative of anything. On those grounds I suggest that "making the multiplication explicit" is actually more misleading than helpful. -- Leland McInnes (talk) 17:07, 10 January 2008 (UTC)

Thank you for your comments. The argument given also applies to the common notation

${\displaystyle f(x)={{dy} \over {dx}}}$

where the division sign is explicite even if the numerator and denominator are formal variables rather than numbers. Formal multiplication by dx gives

${\displaystyle dy=f(x)\cdot dx}$

and formal integration gives

${\displaystyle y=\int f(x)\cdot dx.}$

The link between f(x) and dx is that of multiplication, just as 2x means 2·x, even if x is a formal varable. But I rest my case. Bo Jacoby (talk) 22:21, 15 January 2008 (UTC).

Ultimately you are arguing simply from notational convention rather than how these concepts are formally defined in modern mathematics. The notation is simply a holdover from original Leibnizian notation back when dx really did represent an infintesimal and the sort of manipulations you are using were considered quite normal. Modern definitions of integral and differential avoid such things (are you multiplying by a measure in Lebesgue integrals? What does that even mean? How does it relate to differentials?) but retain the now anachronistic notation out of tradition. Perpetuating the misconceptions involved in infintesimals by taking the notation at face value and even emphasising the anachronistic aspects is, I feel, a bad idea. -- Leland McInnes (talk) 15:03, 16 January 2008 (UTC)

Regardless of whether f(x)dx actually represents multiplication, I'm a little surprised that no one has yet mentioned the possible confusion that may arise with the line integral. It's better to have ambiguity-free notation than notation which will be a potential source of confusion. Besides, the f(x)dx notation is indeed conventional throughout mathematics and its applications, so I believe the Wikipedia policy is to avoid neologisms in this regard. Silly rabbit (talk) 17:03, 16 January 2008 (UTC)

Well, ladies and/or gentlemen, notational convenience is the same thing as algebra, and there is nothing wrong in that. I was surprised that the disagreement was about the multiplication itself, rather than about the multiplication sign. The simpler introduction to differentiation and integration is to begin with polynomials. If x is a formal variable, then so is dx. The rules of algebra are d(x+y)=dx+dy and d(x·y)= x·dy+dx·y. These rules are sufficient for deriving the rules for differentiating a formal power series and for solving differential equations. The rules of interpretation is that if dx is not zero, then x is neither constant, nor maximum, nor minimum. The interpretation in terms of limits, and the Riemann and Lebesque integrals, are not needed for quite a while. The beginner needs a break. Wikipedia should explain, rather than just repeat unintelligibly advanced stuff. Bo Jacoby (talk) 15:13, 20 January 2008 (UTC).

If you want to explain, then explain: the introduction section was meant to provide a place for beginners to get a grasp of what was going on without (and prior to) dealing with the technicalities. Abusing notation with something that is ambiguous (as Silly rabbit points out), nonstandard, and technically wrong for the majority of the article, is not explaining. Having a paragraph in the introduction explaining how things can naively be viewed in the terms you describe (possibly with a link to synthetic differential geometry or smooth infinitesimal analysis for those who would like to see things formally done in such a manner) would be explaining, and that most certainly doesn't require any change of notation anywhere in the article except in the paragraph where you wish to do your explaining. If you wish to actually add an explanation rather than just messing up notation then please, have at it, I will welcome such material for beginners and will even help improve it if I have time. -- Leland McInnes (talk) 16:31, 21 January 2008 (UTC)

The omission of an explicite multiplication sign is widespread. That does not mean that the multiplication sign is a neologism. The article on polynomial omits the multiplication sign, but that does not mean that multiplication is not intended, for example 2xy² means 2·x·y². Don't you all agree on that? Omitting the multiplication sign makes no harm until juxtaposition means something else than multiplication. Then confusion appears. For example (f+g)(x) = f(x)+g(x) defines the sum of two functions, f and g. Here the juxtaposition (f+g)(x) does not mean the multiplication (f+g)·(x). Mathematicians don't mind very much, because the parenthesis around (x) indicate that x is argument to a function, but parentheses have other meanings. In the expression (f+g)(x+y) it is less clear whether the interpretation f(x+y)+g(x+y) or (f+g)·(x+y) is intended. Have for example a look on the articles catenary and gamma function and identify which juxtapositions in the formulas indicate multiplications and which ones do not, and why. Explicite multiplication signs sure would help a lot. So, omission of multiplication signs in formulas may be polite to the author, but it is rude to the reader. Regarding integration there is no doubt that Leibnitz intended a multiplication between the function value f(x) and the differential dx: The differential of the area bounded by the x-axis, the y-axis, the curve y=f(x), and the vertical line at x, is the height f(x) times the base dx. The difficulties in interpreting the differentials have historically lead to tricky definitions for derivation and integration, but the algebraic axiomatic approach avoids these complications. You do not need to know what a differential is, as long as you can use it correctly in computations, just as you do not need to know what −3 means, except that is solves the equation x+3=0. Bo Jacoby (talk) 02:13, 22 January 2008 (UTC).

I agree that 2xy² means 2·x·y², the problem here is that I do not agree that ${\displaystyle \int f(x)\,dx}$ means ${\displaystyle \int f(x)\cdot dx}$, not least because I don't know what ${\displaystyle \int f(x)\cdot dx}$ means (are we talking a dot product in a line integral, are we using formal infinitesimals, are you simply defining your own integral?). Also note that, despite the fact that 2xy² means 2·x·y², we don't go around inserting multiplication signs on every page involving algebra; instead we follow the standard conventions used in maths texts. You can bemoan the context sensitive nature of mathematics often overly compact notational conventions, but they are the notational conventions and as a tertiary source we are supposed to follow the notational conventions not buck them with our own ideas. In the end it doesn't matter what Leibniz intended with the notation; his definitions were shown to be wholly inadequate, and integrals simply do not mean what Leibniz had in mind anymore. The fact that we still use the notation is, as I said earlier, just an anachronism (and indeed, some modern alternative notations that haven't caught on widely enough to be included or used in this article do away with such things). Your own personal ideas of how integrals should be defined are irrelevant here; we only care about how they are defined in the mainstream literature, and that means Riemannian, Lebesgue, and Henstock-Kurzweil for the most part... none of which allow for your "it's really a multiplication" because under those definitions it really isn't. -- Leland McInnes (talk) 02:40, 22 January 2008 (UTC)

Dear Lenand McInnes. Yes, we are talking formal infinitesimals. No, I am not defining my own integral. If

${\displaystyle \int 2x\,dx}$

does not mean

${\displaystyle \int 2\cdot x\cdot dx,}$

then you cannot deduce that

${\displaystyle \int 2x\,dx=2\int x\,dx,}$

because then you cannot rely on the distributive rule of multiplication:

${\displaystyle \int (2\cdot x)\cdot dx=\int 2\cdot (x\cdot dx).}$

In the expression for the Riemann sum

${\displaystyle S=\sum _{i=1}^{n}f(y_{i})(x_{i}-x_{i-1})}$

multiplication is obviously implied

${\displaystyle S=\sum _{i=1}^{n}f(y_{i})\cdot (x_{i}-x_{i-1})}$

Also the article Darboux integral defines the upper Darboux sum of ƒ with respect to P:

${\displaystyle U_{f,P}=\sum _{i=1}^{n}M_{i}(x_{i}-x_{i-1}).\,\!}$

Here too juxtaposition means multiplication:

${\displaystyle U_{f,P}=\sum _{i=1}^{n}M_{i}\cdot (x_{i}-x_{i-1}).\,\!}$

Mainstream litterature on Riemann integral, Lebesgue integral and Henstock-Kurzweil integral, as well as the corresponding wikipedia articles, generalizes the elementary Leibnitz integral assuming that the readers are already familiar with elementary high-school algebraic integration. This assumption cannot be made here. I thought that the confusion about the interpretation of juxtaposition was confined to beginners, but now I realize that I was wrong. Bo Jacoby (talk) 11:32, 22 January 2008 (UTC).

FWIW, let me add my voice to the objectors to Bo's proposal. I agree with Leland McInnes, Silly rabbit et al. Writing ${\displaystyle \int f(x)\cdot dx}$ rather than ${\displaystyle \int f(x)dx}$ would be non-standard, potentially confusing, and, in many ways, simply wrong. Gandalf61 (talk) 11:47, 22 January 2008 (UTC)

(Edit Conflict) So this boils down to formal infinitesimals. Before heading to the next obvious question, I'll point out that nonstandard analysis is called nonstandard for a reason. The obvious question is exactly which infinitesimals do you mean? The nilsquare infinitesimals of smooth infinitesimal analysis? Shouldn't we include an explanation of the fact that we are no longer working with the real numbers and we've dropped the law of excluded middle? Or are we using infinitesimals from the hyperreal numbers, which leads us back to nonstandard analysis. All of Riemann integral, Lebesgue integral and Henstock-Kurzweil integral, explicitly avoid infinitesimals (because of the problems they introduce) in favour of limits: multiplication may be implied prior to taking limits, but in the actual integral (after taking limits) it is not; the dx is formal and can (and should) be read as simply saying "with respect to x" (indeed, some mainstream texts freely omit it altogether). If you wish to write a page on Integrals in nonstandard analysis then by all means use your notation there. This page, however, is about integrals as commonly defined, and those definitions do not admit, nor allow, infinitesimals. -- Leland McInnes (talk) 16:05, 22 January 2008 (UTC)

I'm just curious. What does the juxtaposition of f(x) and dx in "f(x)dx" mean if it does not mean multiplication? Is the formula d(2x)=2dx not involving two multiplications? Bo Jacoby (talk) 15:47, 22 January 2008 (UTC).

The formula d(2x)=2dx is not formally rigrous; it's a nice rule of thumb, but modern calculus in any standard formulation doesn't actually admit such expressions. What does the juxtaposition of f(x) and dx in "f(x)dx" mean in Riemann integrals? Absolutely nothing beyond the fact that the integral is to be regarded as "with respect to x". Where no confusion arises you can actually omit the dx entirely at no loss. The dx had meaning when Leibniz created the notation because Leibniz used infiniteismals which have since been banished from standard analysis in favour of limits. We keep the dx kicking around for old times sake, and because it's useful to keep track of things in multivariate cases, but it is a fairly hollow notational point. -- Leland McInnes (talk) 16:05, 22 January 2008 (UTC)

Agreed. ${\displaystyle \int dx}$ is an operator which maps functions to other functions - there is no implied multiplication. ${\displaystyle {\frac {d}{dx}}}$ is another operator, and there is no implied division. It is correct to write ${\displaystyle {\frac {d}{dx}}(2x)=2}$, but it is formally incorrect (although often done when among friends as a harmless enough shortcut) to write ${\displaystyle d(2x)=2(dx)}$. Gandalf61 (talk) 16:17, 22 January 2008 (UTC)

Thank you for answering. I understand that you consider "d(2x) = 2dx" illegitimate while "d(2x)/dt = 2dx/dt" is legitimate, meaning "d(2·x)/dt=2·dx/dt", involving two multiplications. Am I right? "∫ dx = ∫ 1dx". Right? "∫ f(x)dx = ∫ (f(x)·1)dx = ∫ f(x)·1dx = ∫ f(x)·dx". Right? Algebraic shortcuts among friends seems to be taboo i WP even if they work. The Riemann integral was supposed to generalize the antiderivative, rather than to restrict the algebraic freedom. Bo Jacoby (talk) 13:56, 23 January 2008 (UTC).

I would have to say that most of your statements above that end with "right?" or "am I right?" are, in fact, wrong. They are valid, at best, as informal rules of thumb and calculational aids, but they don't represent correct mathematics. The symbol ∫ dx denotes, in modern calculus, an operator as Gandalf61 points out; we just have a rather anachronistic notation for it. You may we well write it I and denote the integral of a function f by I(f). Should you wish to kep track of what you're integrating with respect to you can just as well use I_x(f(x)). Note the complete lack of implied multiplication there. The symbol I_x no more denotes the integral of the constant unit function 1, than the symbol for the real valued function f denotes the evaluation of the function at the real number 1. The heart of your argument is a thoroughgoing abuse of notation based on informal calculation aid based rules of thumb. The Riemann integral was supposed to formalize and make rigorous the notion of integral which had, up until then, been based on the informal use an ungrounded theory of infinitesimals and hand-waving arguments. That formalization made integrals rigorous (and thus not subject to the errors that were creeping in due to arguments based on intuitive notions that were, in practice, invalid) specifically by banishing infinitesimals and the very sort of intuitive (but potentially invalid and at the lest misleading) notational abuse in which you are engaged. -- Leland McInnes (talk) 16:19, 23 January 2008 (UTC)

The argument used in Fundamental_theorem_of_calculus#Intuition for writing ${\displaystyle \mathrm {d} x=v(t)\,\mathrm {d} t}$ meaning ${\displaystyle \mathrm {d} x=v(t)\cdot \mathrm {d} t}$ does perhaps involve handwaving, but the differential algebra does not. There is no handwaving about the rules of differentiation. Differentiation d is defined on the polynomial ring Q[x] by introducing another variable, dx, and the rules that d(x)=dx, and that dk=0 when k is a number, and d(X+Y)=dX+dY, and d(X·Y)=dX·Y+X·dY when X and Y are polynomials. This is sufficient in order to specify formal differentiation on power series, and all the entries in lists of integrals follow. Limits are not needed for the algebra, only for the interpretation. This means that d(2·x)=2·dx is an algebraic fact. It does not rely on Riemann integration. Bo Jacoby (talk) 16:54, 23 January 2008 (UTC). PS. The lists of integrals says: ${\displaystyle \int \,dx=x+C.}$ Are you saying that this elementary formula doesn't represent correct mathematics? I politely request you to tell me which ones of my statements above, that end with "right?" or "am I right?", that you consider wrong. The summary statistics, that most of them are wrong, is not sufficient. I am trying to figure out what you guys mean. Using your (nonstandard) notation I_x(f(x)), I note the rule that I_ax(f(x))=I_x(af(x)), showing that a constant factor can be moved between two arguments, which implies multiplication. Bo Jacoby (talk) 11:52, 24 January 2008 (UTC).

Letting new readers of Wikipedia see exactly what they've always been accustomed to seeing in books is not going to confuse them. I.e.

${\displaystyle \int f(x)\,dx}$

is universally standard. If dx is an infinitesimal increment of a vector quantity rather than a scalar, and f(x) is vector-valued, then one sometimes writes

${\displaystyle \int f(x)\cdot dx\,}$

meaning the dot-product. In the latter case, one should of course write it in that way. Michael Hardy (talk) 15:14, 24 January 2008 (UTC)

Somewhere buried in the above discussion, I made exactly the same point. Why is this debate even continuing? Silly rabbit (talk) 15:19, 24 January 2008 (UTC)

See here. Septentrionalis PMAnderson 19:07, 24 January 2008 (UTC)

what about an open domain of a specific integral while the integral is defined by the domain?

how big is the difference between the integral on a closed domain vs open domain? 132.72.45.190 (talk) 14:58, 15 January 2008 (UTC)

Dear Anonymous,

The difference between the integral on U and the integral on the closure of U may be arbitrarily large. See Smith-Volterra-Cantor set for the reason.

Sincerely,

Loisel (talk) 17:52, 23 January 2008 (UTC)

This concept is not discussed in the article. I met it in Terence Tao's article on differential forms. —Preceding unsigned comment added by Randomblue (talk • contribs) 19:00, 9 February 2008 (UTC)

shouldn't this be merged with Antiderivative? Professor Calculus (talk) 00:30, 16 March 2008 (UTC)

In short, no, since, very broadly speaking, an antiderivative is a function, and an integral is a value. Although antiderivatives can and are used to compute integrals, since if F is the antiderivative of a real-valued continuous function f then ${\displaystyle \scriptstyle \int _{a}^{b}f(x)\,dx=F(b)-F(a)}$ by the fundamental theorem of calculus. But they are not the same concept, so there is therefore no need to merge. Xantharius (talk) 02:31, 16 March 2008 (UTC)

I also feel that the two articles should remain separate. Some of the more pedantic analysts maintain a firm distinction between the two notions: see for instance the first volume of Dieudonne's treatise. However, the article on Antiderivatives is rather limited in scope at the moment. I would support an effort to expand it to include other more general notions of integral primitives going beyond the usual ones familiar from calculus. (E.g., variational primitives, primitives of the Darboux derivative, etc.) silly rabbit (talk) 14:33, 16 March 2008 (UTC)

I agree with the separation, for reasons stated earlier. Antiderivative and integral are two different things, although related. Oleg Alexandrov (talk) 15:48, 16 March 2008 (UTC)

It's the Calculus, not Calculus when referring to the subject of the Calculus. Every instance of the word calculus in this article would seem better if the article the preceded the word Calculus.

Yes, talk to some English major for confirmation of proper English usage in reference to the word calculus.

Mergatroidal (talk) 23:54, 22 March 2008 (UTC)

There is a discussion of this very topic on the Calculus talk page and in its archives. Although the point did not seem to be resolved, it was pointed out that different countries do have different usages. Like other edits which result in a whole host of changes being made, editors should refrain from making such changes unless there is very bad inconsistency in the article, or unless there is consensus to do so. I think the latter of these will be hard to achieve, so perhaps it would be better to leave the article as it current stands with regard to the definite article. Xantharius (talk) 04:19, 23 March 2008 (UTC)

I agree that a large number of edits should be discouraged unless there is a clear consensus. At the calculus talk page, Arcfrk raised the interesting point that the definite article only makes sense when referring to "the calculus of infinitesimals", for otherwise there are many calculi from which to choose (c.f. "the calculus of variations"). silly rabbit (talk) 13:44, 23 March 2008 (UTC)

The Calculus refers to the entire set of mathematical systems employed to determine change. And then there can be reference to a specific branch of the Calculus: the calculus of derivatives, the calculus of integrals, etc. I suppose it's easier to lay back and say who cares and drop the superfluous words, but if one wants to project the character of being precise, of being more precise with their words (hey professor, this is you I'm talking to ...!), wouldn't it appear appropriate in a classroom, let's say, to refer to the Calculus in this way? All these opinions on proper usage and yet no one knows what's correct? Hogwash. Put your foot down and be the first. Stick out (like a sore thumb.) Think different. Do it. (:~}

Would you be so cavalier in speech to such a stellar mind as this? Mergatroidal (talk) 01:55, 25 March 2008 (UTC)

I don't think that this is such an issue of proper usage, and more an issue of what is going to happen when someone goes through an article that has remained stable for some time on this very point and then makes wholesale changes throughout the article. You might have strong opinions as to whether "A.D." or "C.E." is better or more academic in a particular article, for example, but wholesale changes to make an article go one way or the other are not permitted on Wikipedia. I think this falls into the same category. The main point is that when there is no consensus on a point such as this, and the article is consistent, then changes throughout an article should not be made. If it were inconsistent I'd say go for it; however, the usage on this page is to have no definite article, and thus that is probably the way it should remain unless there are wide calls for it to change. Xantharius (talk) 18:25, 25 March 2008 (UTC)

If everyone writes pistachio, you can argue that this is wrong and that it is an Italian word and should be spelled pistacchio, but, well, that is not how it is usually written in English texts, including dictionaries and names of organizations of pistac(c)hio growers. At some point it stops being wrong; my automated spelling corrector in fact objects to pistacchio and wants to know if I perhaps meant pistachio.

When (the) Calculus was introduced, texts were in Latin, and it was called calculus differentialis and calculus integralis, or calculus infinitesimalis. Latin has no definite or indefinite articles, so "calculus", "a calculus" and "the calculus" are all the same in Latin. But in English texts, once English came into common use for mathematical texts, this was translated as "the differential calculus" and "the integral calculus", together as a unified subject "the differential and integral calculus", or "the calculus of infinitesimals". That is a bit long for everyday use, and so it was shortened to "the calculus" in contexts where one might assume it was understood which calculus. So indeed, historically it is the calculus. However, in common use this is the only calculus known, and the article has since been dropped, also for "differential calculus" and "integral calculus". That is also what you find now in dictionaries and textbooks.. In Wikipedia, we follow common usage. We do not attempt to replace what is by what ought to.  --Lambiam 00:14, 26 March 2008 (UTC)

Lambiam, your discourse convinced me. In common usage among those who find the calculus just another one of oh so many mundane things that exist in life, and if one is not inclined to put on airs to the average man, it would seem stuffy, pedantic to employ the article the when referring to the calculus. Though if one were to mingle with other fans of mathematics, let's say at a party with other mathematicians, slipping in the article in conversation when casually referring to the calculus would seem appropriate and understood to convey the personal esthetic sense of appreciation one has for the beauty of what the calculus is all about. Personally I have recognized total stranger's perceptions for the grandeur of what the calculus is all about when they used the article of description in conversation, and isn't this sense what words are supposed to do: to convey one's thoughts and feelings? It's all context, and attitude. The common Wikipedia author speaks to the common man. Most Wikipedia authors are not elitist, and that is a bit of unintended sarcasm. I suppose let's not strive for the ideal inside Wikipedia articles, and appear snobby, elitist.

"There goes the King!" or "There goes that King guy." Mergatroidal (talk) 22:17, 12 April 2008 (UTC)

As a professional mathematician, allow me to assert that I have never, ever, heard another mathematician refer to "the calculus" at one of our secret cabalistic get-togethers meetings. Funny post though. Cheers, silly rabbit (talk) 00:10, 13 April 2008 (UTC)

— While driving a taxi in New York, and I believe the conversation was about the history of mathematics, and from the back seat a jocular, "... the calculus" was uttered, and not so much to correct my second or third instance of use of the word calculus, instead the passenger was intending to impress upon me that the calculus is a remarkable accomplishment of the human mind. Other than what the Greeks accomplished, the calculus could be put on a pedestal. I am not that versed in mathematics, though I wonder if Silly rabbit could name five other mathematical accomplishments as remarkable, and on par with the calculus? The practicality of which is par excellence.Mergatroidal (talk) 23:25, 21 April 2008 (UTC)

The fundamental theorem of calculus guarantees that once an antiderivative is known, a definite integral can be computed. I therefore see no need for this edit, and find the weasel words unnecessary and misleading. I'm willing to be swayed by a detailed and convincing rationale for the edit. siℓℓy rabbit (talk) 00:53, 24 May 2008 (UTC)

I'm guessing the rationale is that given a suitably unpleasant/obscure/nasty function as the antiderivative, actual computation of a numeric answer for the definite integral needn't be easy (or even technically possible). In general though, I feel that the edit was really being excessively cagey; an answer for the definite integral can be given, if not strictly computed, and there really isn't a need to split such hairs at this point. -- Leland McInnes (talk) 02:57, 24 May 2008 (UTC)

It is interesting that the IP of the anon editor making the initial edit (with edit summary It is not always easy to compute a definite integral even if the indefinite integral is known. This is especially true if the input has parameters.) is registered as belonging to Wolfram Research. I think there may be different perceptions of what is meant by computing a definite integral. If the meaning is interpreted as "producing a good approximation of its numerical value", then the present formulation with "easy" is too facile. What about giving the precise statement?:

The principles of integration were formulated by Isaac Newton and Gottfried Leibniz in the late seventeenth century. Through the fundamental theorem of calculus, which they independently developed, integration is connected with differentiation: if f is a continuous real-valued function defined on a closed interval [a, b], then, once an antiderivative F of f is known, the definite integral of f over that interval is given by:

${\displaystyle \int _{a}^{b}f(x)\,dx=F(b)-F(a).}$

 --Lambiam 16:27, 25 May 2008 (UTC)

I have implemented your suggestion. At least this obviates any argument about the meaning of the word "compute". Interestingly, it appears that the Wolfram IP was concerned more with issues of convergence of the integral. Indeed, if f is not continuous, then the statement is potentially problematic. Of course, saying "in principle" doesn't make this any clearer, nor does it fix the problem. Your suggestion is a much better solution, in my opinion. siℓℓy rabbit (talk) 17:25, 27 May 2008 (UTC)

Someone had apparently a different opinion.  --Lambiam 05:13, 28 May 2008 (UTC)

Oh, whatever. I think I screwed up the edit, and Oleg is reacting to that. But he's probably also right that a detailed discussion of the fundamental theorem is out of place. siℓℓy rabbit (talk) 11:47, 28 May 2008 (UTC)

Ah, I see I reverted more than I wanted. What I did not want in was this, since there it is not the right place to discuss the fundamental theorem of calculus, and it is not stated what the assumptions should be for the theorem to hold. I see I had also reverted some other text, which I now put back. Silly rabbit, which was your intended edit? Oleg Alexandrov (talk) 15:39, 28 May 2008 (UTC)

I'm good and confused. But I'm also just fine with the lead as it is now. I was also fine with it then.  ;-) siℓℓy rabbit (talk) 15:47, 28 May 2008 (UTC)

OK then, you guys can do here whatever you think is right. :) Oleg Alexandrov (talk) 04:50, 29 May 2008 (UTC)

59.103.25.113 (talk) 13:30, 15 August 2008 (UTC) By. Asad S. Yousaf Dated 15th Aug 2008 I was asked to place links to my Area Applets on Talk page rather than on Main page for Integrals. As definite integrals are used to approximate Area under and between curves, Volumes of Solids of Rotation, Length of a Plane Curves, Surface Area, Centroid, and so on. I had implemented online demonstration Java applets that illustrate the concepts just cited. So far I have developed 20 applets, and more can be expected. Yet I realized, putting up a demonstration program on web page without elaboration of the underlying concept served no purpose. So my applet pages are being reworked to include discussion of the topic along with presentation of the applet. So far Area under and between curves and Arc Length pages have been updated. Since Wiki is home to many Calculus topics, I thought your viewers may find my tools useful if they can interact with them to have a Visual representation of such concepts. Allow me to mention the the links to Area applet pages. Area under a Curve Applet is viewable at Area between Curves Applet is viewable at I am all ears to your feedback

I took a quick look at the "Area under a Curve Applet". Some feedback:

• Your notes only cover positive functions, and don't distinguish between integral and (unsigned) area. If you put a function that changes sign into the applet, then it approximates the area between the curve and the x axis i.e. it approximates the integral of |f(x)|, not the integral of f(x). So, for example, f(x)=x between -3 and 3 gives an area of 8.99. I think this is confusing.

• Accuracy is not very good ! One example in point above; another example is f(x)=x² between -3 and 3, where applet gives an area of 17.97 - should be 18.
• Applet doesn't handle vertical asymptotes. f(x)=1/sqrt(x) between 0 and 1 gives an area of "9.22337...E16" - should be 2.

I think this applet has too many problems to be useful to Wikipedia readers in its present form, but keep up the good work. Gandalf61 (talk) 14:49, 15 August 2008 (UTC)

59.103.25.176 (talk) 15:57, 15 August 2008 (UTC)Asad S. Yousaf I have fixed the first two problems you pointed out.

It now uses f(x) rather than |f(x)|

It rounds the answer to a single decimal place, so you should get 18 rather than 17.97

On the third point, for a point where function has a vertical asymptotes do I simply skip the value. Let me know so I can fix this. When I typed in 0.001 to 1 for (x)=1/sqrt(x), it now says 2

59.103.27.86 (talk) 18:26, 15 August 2008 (UTC)Asad S. Yousaf

I need one of yours help on this - A link to my ArcLength page was removed from Wiki ArcLength page by an editor "OdEdSchramm" citing the reason "appears to be spam". I don't think I SPAM anymore than I like HAM. So if one of you here can help me restore the link on Wiki ArcLength page. Thanks

That discussion belongs more properly at Talk:Arc length, but doesn't your website also serve a commercial purpose? See External links: Advertising and conflicts of interest. Even disregarding that, the link does not provide a unique and valuable resource, and basically the same criticism as for the integral applet applies: the results are not very accurate. For example, the arc length of sqrt(100^2-x^2) for x from -100 to 100 is given as Arc Length = 311.8 instead of 314.16 (100π).  --Lambiam 09:19, 20 August 2008 (UTC)

In the introduction, I just made a change:

is defined to be the signed area of the region in the xy-plane bounded by the graph of ƒ, the x-axis, and the vertical lines x = a and x = b.

This doesn't mention how to define the integral (which is beyond the scope of the intro I think), but is it accurate? I'm also a bit worried that it doesn't mention that this area may not exist, but I don't want laden the intro with technical details, so any ideas? Cheers, Ben (talk) 00:56, 16 December 2008 (UTC)

I had the same thought. But I think your wording is best. To be 100% accurate, we should say that the signed area under the graph is defined to be the integral (not the other way around), and the integral is defined by...[whatever]. Nevertheless, the earlier phrasing "...is equal to..." leaves an uncomfortable ambiguity, and the present wording at least captures the general idea of how one tries to define the integral. Indeed, one defines the integral precisely so that this is true, whenever it makes sense. siℓℓy rabbit (talk) 01:08, 16 December 2008 (UTC)

I'm concerned about some of the notation in this article. It's in the Introduction section.

${\displaystyle \int _{0}^{1}{\sqrt {x}}\,dx=\int _{0}^{1}x^{\frac {1}{2}}\,dx=\int _{0}^{1}d\left({\textstyle {\frac {2}{3}}}x^{\frac {3}{2}}\right)={\textstyle {\frac {2}{3}}}.}$

I've never seen ${\displaystyle \int _{0}^{1}d\left({\textstyle {\frac {2}{3}}}x^{\frac {3}{2}}\right)}$ used in any of my textbooks or by any of my teachers. Are you sure that this is standard notation? —Preceding unsigned comment added by Metroman (talk • contribs) 06:49, 3 March 2009 (UTC)

Well, have a look at Riemann-Stieltjes integral. Although it is correct, I don't see the mathematical advantage in using such notation in the given context. --PST 09:07, 3 March 2009 (UTC)

There seems to be some confusion in the third bullet point under Linearity in Properties of integral. There is little point in requiring the space V to be locally compact: over non-discrete valued complete fields that requirement forces the space to be finite dimensional. In addition, the discussion and conditions imposed indicate a possible confusion between strong ("Bochner") and weak ("Pettis") integrals. Depending on how deep one wants to go, it would make sense to discuss:

1. Integral with values in a Banach space (here need the completeness assumption, but crucially also the norm (necessary condition for integrability is that the norm of the function be integrable)). This is the strong integral, and is a rather straightforward generalisation of the real and complex-valued integrals.
2. Integral with values in a Hausdorff locally convex space (was this the source of local compactness, or does that come from the domain of integrable functions - Bourbaki style?). Here the integral is defined for scalarly integrable functions (f for which the function x → <v*, f(x)> is integrable for each x* in the topological dual of V. This integral is then a (not necessarily continuous) linear form on the topological dual of V, i.e., belongs to V '*, which is precisely the completion of V equipped with the weak topology. Hence the question of whether the weak integral belongs to V or only in its weak completion, which seems to be alluded to in the present text.

However, I'm not inclined to implement the above changes at that particular point in the article, where they do not properly belong. Would be better to be content making the point there that the various integrals are all linear operators on the (vector) spaces of functions where they are defined. Instead, there should be a short summary section on vector-valued integrals, linking to articles on weak and strong integrals. Stca74 (talk) 21:50, 10 March 2009 (UTC)

I've been reading a little about different definitions of the integral, and a couple of books mention the "Cauchy Integral" which was formulated before the Riemann Integral and is in fact a special case of the latter where the "tag" of each interval in the partition is chosen to be the left endpoint of the interval. I notice that Wikipedia (and seemingly most other online sources from a quick google) doesn't mention it at all, and Riemann Integral even goes as far to say "the Riemann integral, created by Bernhard Riemann, was the first rigorous definition of the integral of a function on an interval." when it seems that Riemann merely generalized Cauchy's integral. Cauchy integral is also a redirect to Cauchy's integral theorem.

I feel like it should be included for completeness, at least from a historical perspective if nothing else. I'm no expert by any means, but I might have a go at making an article. It seems fairly odd that it is not referred to on WP at all so I'm not sure whether to just plow ahead and make an article (and change the Integral and Riemann Integral accordingly). It seems it would be fairly straightforward to include since the definitions are so similar to the Riemann.

So basically I'm just wondering if anyone objects to including this integral in WP, or knows anything about it. (also posting this at Talk:Riemann integral) slimeknight (talk) —Preceding undated comment added 02:38, 25 November 2009 (UTC).

The portion that introduces the idea of the integral, when evaluating, simply goes to the integral to F(1) - F(0), which then evaluates to ²/₃. Should it be mentioned that ∫x^1/2dx = ²/₃*x^3/2? MathMaven (talk) 16:10, 6 March 2010 (UTC)

A request for comments has been filed concerning the conduct of Jagged 85 (talk · contribs). That's an old and archived RfC, but the point is still valid. Jagged 85 is one of the main contributors to Wikipedia (over 67,000 edits, he's ranked 198 in the number of edits), and practically all of his edits have to do with Islamic science, technology and philosophy. This editor has persistently misused sources here over several years. This editor's contributions are always well provided with citations, but examination of these sources often reveals either a blatant misrepresentation of those sources or a selective interpretation, going beyond any reasonable interpretation of the authors' intent. I searched the page history, and found 2 edits by Jagged 85 in August 2008. Tobby72 (talk) 20:56, 11 June 2010 (UTC)

I followed a link here, but square-integrable is not defined. Nor is it obvious what it means.

According to Wolfram, f(x) is square integrable if the integral of the |f(x)|^2 dx from -infinity to +infinity is finite.

201.229.37.2 (talk) 11:44, 27 August 2010 (UTC)

Yes, more precisely, f(x) is square integrable on an interval (or more general set) if the integral of |f(x)|^2 dx on that set is finite. I changed the redirect . The target mentions also that in special cases we have to distinguish which definition of integral we use.--Patrick (talk) 12:54, 27 August 2010 (UTC)

This article speaks to the layman and gives a simple example early. Only later in the article does it get to more technical issues. This is helpful to the general population of readers. I wish more Wiki authors followed this example when writing about complex math and science issues. —Preceding unsigned comment added by 99.147.240.11 (talk) 20:10, 3 September 2010 (UTC)

The inequalities section is great, but for the benefit of non-mathematical scientists it may be worth a passing mention whether integrals preserve strict inequalities. That is, if f(x) < g(x) for all x in [a,b], then:

${\displaystyle \int _{a}^{b}f(x)dx<\int _{a}^{b}g(x)dx.}$

For a such a subtle change I have actually found this useful in applications, so I think it is worth putting in the article. However not having studied Lebesgue integration formally I'm not 100% sure if it's always true, so I put it up for discussion. 188.220.4.91 (talk) 21:55, 11 March 2011 (UTC)

This fact is true. Suppose that f < g. Then 0 < g − f. Hence 0 < ∫(g − f)dx, and it follows that ∫f dx < ∫g dx. Ozob (talk) 15:19, 12 March 2011 (UTC)

The area of a region is increasing by a rate of ${\displaystyle f(x)}$: which means the vertical distance between (x,0) and (x,f(x)). This represents dA/dx=f(x).

Then integrate the area function A(x), which is the reverse of differentiation and we get the area of a function bounded by a curve and the x-axis. Am I right? Garygoh884 (talk) 01:07, 22 May 2011 (UTC)

I am not sure what you are saying, but I think you may find calculus of variations helpful. Ozob (talk) 03:10, 22 May 2011 (UTC)

I am sorry, but this article fails as it does not say in simple terms what an integral is from the beginning.

What you need to do is have a very, very simple definition at the beginning and then work up to the technical stuff later. This enables people to understand at the beginning roughly what it is. If they need to know more, it also informs this learning process and is altogether a good thing.

Can someone who does understand the subject do this? BTW contrasting this with differentation does not help as us maths thickos don't know what that is either (which is why we are here in the first place....) — Preceding unsigned comment added by 131.111.27.50 (talk • contribs)

There is an informal definition of a definite integral at the end of the first paragraph. It says that the definite integral of a function ƒ(x) between the limits a and b is "the net signed area of the region in the xy-plane bounded by the graph of ƒ, the x-axis, and the vertical lines x = a and x = b". There is also a diagram illusrating this definition. It is difficult to provide a simpler definition which is at the same time reasonably accurate. Which part of this definition do you not understand ? Gandalf61 (talk) 10:47, 15 June 2011 (UTC)

Most of it - I know what most of the words mean (although I hadn't come across the phrase "net signed area" before) but I am not sure what to make of their combination. An example from the real world might be an idea; eg the integral of (example) is (what?), can be found (how?) and is measured in (what?), this can be seen on the graph (illustration). A more general definition is (something like yours above). — Preceding unsigned comment added by 131.111.27.50 (talk) 16:37, 20 June 2011 (UTC)

There is quite an extensive worked example, with a diagram, in the Introduction section. Gandalf61 (talk) 18:18, 20 June 2011 (UTC)

The following text in the introduction to this article got me wondering:

[...] the definite integral [...] is defined informally to be the net signed area of the region in the xy-plane bounded by the graph of ƒ, the x-axis, and the vertical lines x = a and x = b.

Given a function ${\displaystyle f:\mathbb {R} \rightarrow \mathbb {R} _{+},\ f\in C^{\infty }\ }$, is "the area under the graph" not also the formal definition? As far as I know, the motivation behind both the Riemann and Lebesgue integrals is measuring areas under curves and irregular volumes in a meaningful way. Furthermore, the ways I've seen Lebesgue/Riemann integrals developed and motivated usually emphasizes that definitions are consistent with areas or volumes.

Bottom line: let ${\displaystyle f:x\mapsto f(x)\ }$ have the properties as above. Is "the area under the graph of ${\displaystyle f\ }$ between ${\displaystyle a}$ and ${\displaystyle b}$" not a valid, formal definition of ${\displaystyle \int _{a}^{b}f(x)dx\ }$? If not, why not?

Cheers! Trolle3000 [talk] 05:32, 23 June 2011 (UTC)

A single definite integral is *always* the area between the graph and the axis wrt integration.

Another way you could think of the definite integral is as a product of two averages: one is the average length of the infinitely many vertical lines in the region and the other is the interval width (infinitely many horizontal lines in a rectangle representing the area of the region).

A hardly known fact is that all integrals are indeed *line* or *path* integrals. As for Lebesgue theory - it is not required in any form or shape.

71.132.128.219 (talk) 21:16, 23 August 2011 (UTC)

I can see some problems with using "the area under the graph" as part of a formal definition of integration:

1. We only have a direct (i.e. non-calculus) way of calculating area for simple geometric shapes such as rectangles and triangles. To formally define the concept of the "area" enclosed by a general curve, you would have to approximate the region under the curve by a set of rectangles (or some other simple shape), add up the areas of the rectangles, then see if there is a limit as the width of these rectangles tends to zero. In effect, this is the formal definition of the Riemann integral - so this just introduces "area" as an intermediate concept into the standard definition.
2. Using "the area under the graph" as a formal definition for both Riemann and Lebesgue integrals does not explain why there are functions that are integrable under the Lebesgue definition but not under the Riemann definition - unles you say that "area" means different things in the two definitions, which is somewhat confusing.
3. It is not obvious how "the area under the graph" definition generalises to related concepts such as arc length integrals and contour integrals. Gandalf61 (talk) 08:05, 23 June 2011 (UTC)

I realize this simple definition will not hold when talking about integrals from ${\displaystyle -\infty }$ to ${\displaystyle \infty }$ or when we're dealing with limits of sequences of functions.

However, I think we can agree that a simple, closed, non-pathological curve encloses some well-defined property that could be called area, and that a curve homeomorphic to the real line has some well-defined property that could be called arc length? After all, those properties can be measured with either

1. a piece of string and a ruler, or
2. cardboard, scissors and good kitchen scales.

As long as we're talking about functions on compact intervals in ${\displaystyle \mathbb {R} ^{3}}$, why shouldn't we be able to assign physical meaning to the integrals? Trolle3000 [talk] 07:10, 23 June 2011 (UTC)

That's what the informal bit in the lead is about. It is no way to go around a formal definition of an integral though, there is more than one definition and the area would be defined by the integral so it is a bit of a tautology. For instance in one definition the area can't be defined if all the points on the graph are one except for the rationals where it is zero. In others there's problems with the function going off to infinity or dealing with a path integral round a point in the complex plane. What on earth is the area under a complex number function? Dmcq (talk) 08:55, 23 June 2011 (UTC)

What I think the original poster wants to use is the following statement: If S is the region under the curve and μ is Lebesgue measure, then ${\displaystyle \mu (S)=\int _{a}^{b}f(x)\,dx}$. That's true. But the usual way of proving it is to prove that both sides are equal to ${\displaystyle \iint _{S}d\mu }$. That points out a philosophical difficulty with this approach, namely that you have to go up one dimension. So if you want to define surface integrals, then you interpret them as volumes of appropriate regions (specifically, regions in the normal bundle to the surface). If you want to define volume integrals, then you have to measure hypervolumes. Etc. One can do this in principle—Lebesgue n-measure is defined for any n, so the definition is not circular—but it would be, I think, messier than the usual treatment. And one would still want to have a description of which functions are integrable, limit theorems, and so on, and I think that would be more difficult in this framework because the definition of the integral is so indirect. Ozob (talk) 10:36, 23 June 2011 (UTC)

I guess the lead could mention that measure is a mathematical generalization of the concept of area and that might make some of the rest more accessible. Dmcq (talk) 17:12, 23 June 2011 (UTC)

I think what really bugs me is the word "informally" - it leaves the reader thinking that there is something more to it than area, and there really isn't - apart from all the math, of course! In the article about integrals at Mathworld it says:

An integral is a mathematical object that can be interpreted as an area or a generalization of area.

I like that statement. It is precise, and doesn't leave the reader wanting information. So I propose we edit the text in this article to:

Given a function ƒ of a real variable x and an interval [a, b] of the real line, the definite integral [...] can be interpreted as the net signed area of the region in the xy-plane bounded by the graph of ƒ, the x-axis, and the vertical lines x = a and x = b.

What do you say? Trolle3000 [talk] 17:27, 23 June 2011 (UTC)

Mathworld tends to be informal anyway. Even so you left one very important thing out from what Mathworld said - 'or generalization of area'. Without that you are left with the informal bit. There is more to it than area. When a mathematician generalizes a teacup can turn into a doughnut. Dmcq (talk) 17:51, 23 June 2011 (UTC)

@Dmcq, what you write is one of the things I love about math, but also the reason I think the formal/informal discussion should be left out of the introduction - no need to complicate things further. When dealing with a function ${\displaystyle f\in C^{\infty },f:[a,b]\in \mathbb {R} \rightarrow \mathbb {R} _{+}}$ and an integral ${\displaystyle \int _{a}^{b}f(x)dx}$, why make it messier than necessary? In this case, how can you interpret the integral if not as the area below the graph? And why not tell that out loud? Trolle3000 [talk] 22:47, 23 June 2011 (UTC)

If f represents velocity, then I am much more inclined to interpret its integral as displacement. To my mind, displacement is a more accurate description of an integral than area. (For example, the interpretation as displacement is true even if f is not non-negative.) But even that is only valid in the one-dimensional case. Ozob (talk) 23:45, 23 June 2011 (UTC)

Now we are talking physics. If "f(t)" represents (non-negative) acceleration and "t" represents time, then the integral as well as the area below the graph represents speed. If "f(s)" represents (non-negative) force and "s" represents displacement, then both the integral the area below the graph represents work. But we are still dealing with an area, only with other units than "length x length" Trolle3000 [talk] 23:57, 23 June 2011 (UTC)

No, once you make those interpretations, the integral is no longer an area. The integral is only an area if f is interpreted as height above the x-axis. Ozob (talk) 10:27, 24 June 2011 (UTC)

I have removed the paragraph

That same century, the Indian mathematician Aryabhata used a similar method in order to find the volume of a cube.^{[2][verification needed]}

1. [1]

   http://wood.mendelu.cz/math/maw/integral/integral.php
2. [2]

   Victor J. Katz (1995), "Ideas of Calculus in Islam and India", Mathematics Magazine 68 (3): 163-174 [165]

since it has little in common with what the cited article states:

The formulas for the sums of the squares and cubes were stated even earlier. The one for squares was stated by Archimedes around 250 B.C. in connection with his quadrature of the parabola, while the one for cubes, although it was probably known to the Greeks, was first explicitly written down by Aryabhata in India around 500

Sasha (talk) 22:57, 2 January 2012 (UTC)

This article has been edited by a user who is known to have misused sources to unduly promote certain views edits (see WP:Jagged 85 cleanup). I searched the page history, and found 3 edits by Jagged 85. Examination of the sources used by this editor often reveals that the sources have been selectively interpreted or blatantly misrepresented, going beyond any reasonable interpretation of the authors' intent. Tobby72 (talk) 19:22, 19 January 2012 (UTC)

"Area under the curve" redirects here, but this page does not define an AUC in terms of its use in statistics or give the reader an indication of how they should interpret an AUC when they first encounter one. — Preceding unsigned comment added by 145.117.146.70 (talk) 10:33, 14 October 2010‎ (UTC)

integration of cos x/sin²x — Preceding unsigned comment added by 41.221.159.84 (talk) 15:58, 18 February 2012 (UTC)

I'm not totally sure, as the article on the "transport function" is very short, but I'm pretty sure that the "transport function" is NOT a definition of the integral as is stated in this article on integrals. 24.18.97.156 (talk) 01:23, 11 April 2012 (UTC)

At the moment this article tries to cover too much. It might be in order to split it into an article on single-variable, real-valued integration (which could then talk much more about applications of these basic integrals), and a more general article on integration, its history, and a list of types of integration written in summary style. — This, that, and the other (talk) 09:45, 13 May 2012 (UTC)

I think the length and coverage are about right for a top level article like this. There's already summary style happening in each section. I don't find the emphasis on any one topic to be overwhelming. Overall, it's a well-balanced article of an appropriate length. The main deficiency is better citation style. Sławomir Biały (talk) 12:06, 13 May 2012 (UTC)

Hi!

According to my sacred texts, any continuous function on the closed interval [a,b] is Riemann integrable over that interval. Now there exist functions satisfying that condition - hence integrable - but nowhere differentiable. So, forgive me my ignorance, but I take this to mean that the integrated function (although it can't be expressed in a closed form) is differentiable, once. It seems a bit screwy. Have I misunderstood something? In any case, might it be worth mentioning integration and these functions in the article regarding Riemann integration? All the best 85.220.22.139 (talk) 16:13, 28 July 2013 (UTC)

Whoops! My mistake. I was talking about a definite integral which has, of course, a numerical result. I beg yoyr forgiveness, but I still think that nowhere continuous functions deserve a mentions. All the best. 85.220.22.139 (talk) 17:12, 28 July 2013 (UTC)

You might appreciate differentiability class. Ozob (talk) 20:17, 28 July 2013 (UTC)

I am writing here about this edit, whose edit summary reads "Layout/formatting changes and formatting/cleanup templates added. Moved history section to the end of the body and moved an oversized image out of the lead. This page really needs a lead rewrite." My inclination is to revert this edit, since I disagree with everything that it did:

1. The edit removed the image from the lead, with no real justification except to say that it was oversized. It is not an oversized image: in fact its dimensions are quite typical for a lead image.
2. The edit added the template {{lead rewrite}} with the justification "The current lead section lacks sufficient generality to summarize all forms of integrals (and hence, the article)". It may well be true that the current lead does not summarize all forms of integration, however it does support the article as currently written. All aspects of the article are summarized in the lead, roughly in proportion to their prominence in the article.
3. The edit added the template {{too many photos}}. I can't see how this is remotely the case. Many sections have a single image in support, the chief exceptions being the section on Riemann integration which has two, and the long introduction section which has three (that comfortably fit within that section). This does not seem at all to be excessive.
4. Moving the history section to the very end of the article seems to run against the purposes of WP:MTAA. The history is the most accessible portion of the article, so it should be nearer to the top than to the bottom.

--Sławomir Biały (talk) 11:40, 21 October 2013 (UTC)

• That's not everything I did in my edit - you reverted changes that addressed conformity problems with WP:IMAGELOCATION and WP:LAYIM - (see formal definitions section where the line is cut). MOS indicates there are too many images when there's overrun into another section. I put in a temporary fix with the "clear" template, which you then reverted. The ideal fix would be to use a gallery to group the images neatly, not delete images.
• The sandwiched text between the left and right images under Riemann integral is (IMO) the worst MOS flaw/appearance issue on this page. (WP:IMAGELOCATION) If you don't like my fix, you need to do something else to address it.
• The TOC is extending the lead section due to its excessive size and position - the only solution I know of for fixing that is a TOC limit.
• Neither the first definition, nor the first paragraph of the lead, sufficiently define the integral conceptually or mathematically in a general context (which would describe the integral of a map over an arbitrary space). At minimum, it should mention more general integrals (i.e. types of integrals and spaces over which one can integrate) in this paragraph. I do not think it's a good idea to use a mathematical definition of the integral in this general context, because that would be too technical for most readers; however, I think it's absolutely necessary (it's also indicated in WP:LEAD and more specifically in WP:GOODDEF) to adequately describe integration in general terms, not in specific cases. The current lead would be great for an article on Riemann integration on the real line; but, you'll need to explain to me exactly how the first paragraph reflects upon a Lebesgue integral over an arbitrary/general measure space, because I don't see it. I can't think of an adequately general definition/description off the top of my head, but it should answer the question, "What does the integral of a mapping actually represent in practical terms?"
• I think you raised a good point about keeping the history section as the first section though.
• Also, I misread the source code information on my browser when I checked the lead image - I read the default size (420px) instead of the current size (300px) For future reference, a lead image is "oversized" (by policy definition) if it is >300px, per WP:LAYIM. So that was a reasonable thing to revert.

How would you prefer to address these remaining issues? Seppi333 (talk) 17:37, 21 October 2013 (UTC)

I missed some minor formatting changes, but the edit was not adequately summarized. (It would be more helpful to roll this out as a sequence of edits, each with an informative edit summary about precisely what was done rather than relying exclusively on a diff to determine what had changed.) I have fixed the text squashing issue and set the TOC limit to 2.

I don't really follow your point about the lead being too specific. The Lebesgue integral also measures the signed area under the graph of a function, so it's not overly specific to context of the Riemann integral. It would be inappropriate to attempt in the first paragraph to emphasize the general case of an abstract measure space since this is treated only briefly in the body of the article itself. Whether this focus is appropriate is ostensibly a problem with the article, not with the lead. Sławomir Biały (talk) 21:29, 21 October 2013 (UTC)

I'll put in a gallery and add content on the Lebesgue integral once I've finished taking amphetamine to FA status. For the lead, I really just meant the definition or description should encompass that kind of integral over that form of space in addition to a Riemann integral on R. Basically, what's stated doesn't describe the mathematical term "integration" in general - so it's incomplete, not wrong. I'll contact you for your input on addressing this when I'm ready to work on it (assume it isn't fixed before then).

Regards, Seppi333 (talk) 17:25, 22 October 2013 (UTC)

I'm not sure what content you had in mind with respect to the Lebesgue integral: the lead already mentions that way of integration and, moreover, already includes the basic intuition that is common to both Riemann and Lebesgue integration. For readers wishing to know more about these different notions of integration, there are actually separate articles Lebesgue integral and Riemann integral. However, there are many other kinds of integrals: for instance the Denjoy integral, the Henstock–Kurzweil integral, the Daniell integral, the Gelfand–Pettis integral, the Riemann–Stieltjes integral, the Lebesgue–Stieljes integral, and so forth. So it's not at all clear what your ideal lead should look like, nor how you could possibly reference such a lead that encompasses all of these standard generalizations of the usual integral (nor whether such an attempt would actually be an improvement for the typical reader of this article).

While I would enjoy immensely a serious attempt to clarify to another mathematician what the noun "integral" actually means, I should caution that there is ostensibly a plethora of different notions of "integral" that might be of interest to, say, a high school student, an undergraduate major in the sciences, a mathematics major, a graduate student, or a mathematics researcher. This article, as a top-level article on the topic, should probably cater to the least common denominator of this group. The most common intuition is the area under a graph, as the lead already discusses, and this intuition is actually valid for both the Riemann and Lebesgue integrals. A specialist interested in a particular kind of integral should be able to navigate easily to more specialized articles (there are links in the text as well as navboxes and categories) whereas a novice needs a description of the topic that is familiar and easy to understand.

The bottom line is that if you find that our treatment of the Lebesgue integral is lacking, then the appropriate article to edit is Lebesgue integral, not necessarily the main page Integral, just as if you were to feel that we did not adequately address the difference between the Henstock&endash;Kurzweil integral and the Daniell integral, for example, then the appropriate place for that discussion would be on some subordinate article. Sławomir Biały (talk) 00:45, 27 October 2013 (UTC)

Hi Slawomir, I'm still not ready to work on this yet, so I won't be able to really follow up - but the very least that I can say is that when I do work on it, I'd be using WP:RS as is required - my own definition is moot in relation to the topic. I'm well aware that there are many different types of integrals. That's precisely why such a definition in the lead should encompass integration theory in general. Regards, Seppi333 (talk) 01:33, 27 October 2013 (UTC)

In the Terminology and notation section, it says "Some authors use an upright d (that is, dx instead of dx)", when ISO 80000-2-11.16 shows that an upright Roman type is written for the differential. Should the article be changed to reflect this? — Preceding unsigned comment added by 94.9.152.183 (talk) 14:10, 19 July 2015 (UTC)

No. This has been discussed many times before. I have always maintained that an upright d is an error, and others have said that they too prefer an italic d. Additionally, the principle of WP:RETAIN says that we should not make stylistic changes such as this (except to make an article internally consistent). Ozob (talk) 17:53, 19 July 2015 (UTC)

In general Wikipedia goes by usage rather than by standards. I keep on seeing bits being quoted from the ISO 80000 series and disagreeing with what they say, I wonder if no mathematicians were consulted as it says 'to be used in natural sciences and technology'. You'll sometimes see bits in Wikipedia about how things are represented one way in physics and then another way in mathematics. Dmcq (talk) 20:05, 19 July 2015 (UTC)